Before writing, most troubleshooting is a matter of probability. A large fraction of failures can be traced to a single cause. The troubleshooting page is useful, but only if that cause is the leading candidate. If the pages are organized by subsystems, that cause would appear at the seventh position after six checks that were never the cause.
To complete the job, vocabulary has to be considered. Querying search engines is done by individuals as they replicate the error string. The page has to include the string in full, with punctuation, and the error code, as opposed to a neat summary.
Listing the most likely causes for an issue means looking at resolution data and having a debate with an engineer who thinks the rare cause is more interesting. This is an example of making a judgment based on evidence, and it is the most important piece of the resolution decision making.
Models create error codes. When asked to document a failure, these models produce codes that look plausible. They create a log path, and menu items that the product has never issued. The customer searching the real string finds nothing. Fabricated diagnostics are worse than having none at all. This is because people follow them.
Please submit the error catalog and accept some closed tickets with those errors. The author counts what fixed each one and lists the causes in the order of occurrence and issues the checks. Anything a customer cannot resolve is noted to be unsolvable early and to avoid people wasting ten minutes during a server-side outage.
It is better to price it per failure than per product. A single error message spanning a page, which is around 300 words, equates to $30 worth of writing and a 30 cent fee. An issue that branches out to six different causes is likely to require 900 words. Start with the top three errors that most frequently appear in your queue, and measure the results before ordering the remaining errors.
One rate, whatever the format: $10 per 100 words. You are paying for the writer’s time and judgement, so the price scales with the words rather than with a package tier.
| Typical troubleshooting content | Words | Writing | Fee (1%) | You pay |
|---|---|---|---|---|
| Single error message page | 300 | $30 | $0.30 | $30.30 |
| Symptom-to-fix guide | 900 | $90 | $0.90 | $90.90 |
| Full diagnostic tree | 2,000 | $200 | $2 | $202 |
The writer receives 100% of the writing price. Our 1% fee is added on top of it, and 0.5% is donated to tree planting. Full pricing breakdown.
Every draft goes through an editor, then through several AI detectors and a plagiarism check. Anything that does not come back clean is revised or reassigned. You get the reports with the delivery email.
What the piece is for, who reads it, roughly how long it should be, and anything it must include or avoid. Links to your existing material help. If you are unsure, send what you have and the writer will come back with questions.
Sure. The technique is to rank causes according to what actually resolved the returned units, name the components the way the customer views them printed on the device, and be clear on when a fault means the product requires servicing, not another try.
Every time. When three of the top five causes are interface issues, we draft the pages and include a brief note saying so. Sometimes a troubleshooting page is bug documentation, and you should be aware when you are buying one.
Completely. Copyright transfers to you on delivery, with no attribution requirement and no licence back to us. Nothing written for you is resold, repurposed or republished.
Reviews from completed, paid orders in this category.
Thirty articles with the same structure and with the same voice are difficult to develop for a knowledge base. Screenshot placeholders were described with prose instead of with tags. Unfortunately, our designer had to try to find them with some difficulty. This is a small thing, but I plan to put it in the brief next time.
Twenty-two articles on our top ticket drivers. The writer requested redacted transcripts, and then described the problems in the way customers used the language – thermostat will not hold temperature, instead of thermostat calibration. Help center search finally started working.
Eleven articles covering our top ticket drivers. They requested the original ticket text first, not the internal descriptions, and the difference is significant. The articles use the terms customers input into the search box. There’s been a downward shift of about a fifth in the volume of contacts on those topics.
We do not ask you to take the no-AI promise on trust. Each draft is run through these platforms and the reports are attached to your delivery email.
Originality.aiAI detection + plagiarism
GPTZeroAI detection
TurnitinAI detection + similarity
CopyleaksAI detection + plagiarism
Winston AIAI detection
ZeroGPTAI detection
SaplingAI detection
Content at ScaleAI detection
Send the brief, get a human draft back within 3 days. No account, no subscription, no AI.
$10 per 100 words · all of it to the writer · 0.5% to trees